hi guys
is it possible to run an AP on A/B/G mode with both 2.4 and 5.8 one for backbone to server (5.8) and one for hotspot (2.4)

and use this a diplexer such as this to separate signals into two antennas.

thanks

You're talking about two radio cards, merging the antenna's into one tower cable, and then splitting them apart at the top, right?

Two of those duplexers costs less than just putting a second run of LMR up the tower? You're going to be introducing signal loss with each additional connection point also, so that 4 more connections than you'd normally need that would be susceptible to water infiltration and signal loss.

This buys you what?

I don't think MT supports A/B/G on one card. It's A or B/G, not both. Someone please correct my if I'm wrong.

No, I don't think anybody makes a card that can do 2.4G and 5G at the same time, and I'm hoping that wasn't what hellbound was actually asking about.

No, I don't think anybody makes a card that can do 2.4G and 5G at the same time, and I'm hoping that wasn't what hellbound was actually asking about.

you mean you never heard of Super AG?

nobody from mikrotik never heard of it?

just google "Super AG"
combination of 5.8 and 2.4 toghether:

http://www.sparklan.com/product_details.php?prod_id=13

it can boost speed a lot.

and those guys had reason to make that device

any idea?

You're talking about two radio cards, merging the antenna's into one tower cable, and then splitting them apart at the top, right?

Two of those duplexers costs less than just putting a second run of LMR up the tower? You're going to be introducing signal loss with each additional connection point also, so that 4 more connections than you'd normally need that would be susceptible to water infiltration and signal loss.

This buys you what?

no this is not what I'm talking about.

no this is not what I'm talking about.

The device you pointed us to is not a single card that does all three bands, it's an AP with two seperate wireless cards in it, one for each band.

WX-7800A contains two separate wireless connectivity radio transceivers, which support all three popular wireless networking specifications.

I'll ask again, just what is it you're trying to do, since it isn't two cards combined into a single antenna cable and demuxed at the antenna end?

I understand what you are asking now.
First off I dont think that an AP can run A and G at the same time with out two radios connected to two antennas. A client card will scan for A and G APs but connects to one or the other not both.
But lets just say that they can, then your problem would be that MT will force you to pick one or the other. So in MT you can't do it with one card.
But You can install two cards, one for A and one for G. Then you could connect them to two antennas or add the hyperlink Diplexer for two radios to one antenna, as long as the antenna supports A and G.

Did this help or did I miss the point again!?!

FIRST: there is NO such miniPCI card in market that actual DOES 2.4 and 5 GHz at the same time

SECOND: stop dreaming/spamming the forums with useless topics about things that actually DOES NOT EXIST

seriously, i almost fell of chair

i wonder what will be next? a/b/g/turbo/and 900MHz from one parabolic sector antenna?

FIRST: there is NO such miniPCI card in market that actual DOES 2.4 and 5 GHz at the same time

SECOND: stop dreaming/spamming the forums with useless topics about things that actually DOES NOT EXIST

seriously, i almost fell of chair

i wonder what will be next? a/b/g/turbo/and 900MHz from one parabolic sector antenna?

I think he got the point with out your post!

Hellbound,
If your dreaming that's good! Dreamers are the people that make this happen in this world. The rest of us just follow.

actually that is technologically possible (I mean, Hellbounds dream of 1 mPCI with 2 bands)

it will be more like 2 mPCI cards together in one slot.

the drawback would be - if one card is blown by accident (lightning and stuff) the other will be rendered useless also.

as result you will end with greater expenses that you have now.

or create something like RB14, but for mPCI interface

FIRST: there is NO such miniPCI card in market that actual DOES 2.4 and 5 GHz at the same time

SECOND: stop dreaming/spamming the forums with useless topics about things that actually DOES NOT EXIST

seriously, i almost fell of chair

i wonder what will be next? a/b/g/turbo/and 900MHz from one parabolic sector antenna?

it is really shame that your company develop driver mainly for atheros and you have not actually read atheros's suer-g.com completely and never heard of super-AG,
you better look closely to find mini pci that work on SUPER-AG.
if you couldn't find use atheros.com product finder. it might give you some clues.

just for your information check this:

http://www.super-g.com/faqs.html

To be honest with you I did a mistake to chose Mikrotik, and since I have purchased and implemented my network I have no other choice unless I have some extra money to convert.

nice link hellbound, but where is the catch?

super AG mode gives same theoretical speed as if you set atheros based chip R52 in turbo mode (or whatever you call available data throughput to 108Mbit/s air rate)

and i looked and atheros products and find NONE that could provide with one card (integrated or no) 802.11g and 802.11a standard simultaneously

and your idea is bad to end-user - due to fact, that i mentioned before, there are possibility, that one is blown, so you have to change a whole card.

as production of such card would be more expensive, and replacement of such card would me even more pricier.

it would be good for sales but not customer.

and since this is user forum, i can only state my opinion and mine only.

nice link hellbound, but where is the catch?

super AG mode gives same theoretical speed as if you set atheros based chip R52 in turbo mode (or whatever you call available data throughput to 108Mbit/s air rate)

and i looked and atheros products and find NONE that could provide with one card (integrated or no) 802.11g and 802.11a standard simultaneously

and your idea is bad to end-user - due to fact, that i mentioned before, there are possibility, that one is blown, so you have to change a whole card.

as production of such card would be more expensive, and replacement of such card would me even more pricier.

it would be good for sales but not customer.

and since this is user forum, i can only state my opinion and mine only.

Here is some cards that can operate on Super-AG modulation:
they're cheap and easy to find.

I don't quite understand why mikrotik officials in forums are not really
announced by their nickname or logo.
all we know that all mikrotik officials are putting comments in the forum
as it is a planet outside mikrotik broundry.

why do they always have to play "WE-ARE-ALWAYS-RIGHT" game?

when do you really want to officially confirm that you are wrong about
any matter?

Why I have not received any support from mikrotik except vague answers?

even microsoft a multi billion dollar company who has a big reputation in
buggy softwares will dare to announce "MICROSOFT CONFIRM THIS IS
MICROSOFT WINDOWS BUG"

please do not offend anyone in this forum personally or impersonally
when you have lack of knowledge or your other party (like me
does not have good knowledge).


I hope I can receive official answers from mikrotik, I don't really
understand who's the boss there or who make statement and how
shall we deal with this forum?

If i have offended you or anyone else i deeply regret that and ask for forgiveness. I did not mean to do so.

These cars you just posted here DOES support TurboA, turboG modes they really do, here you are correct, but they do not do so at the same time, you cannot set then in turbo-AG mode so they are working in 2.4 GHz and 5GHz.

About forums - these are announced to be user forums, where one user replies to another, and as all we are human beings we can be wrong, all of us.

If i have offended youo so or anyone else

Nope, not offended.

I think I see where the misunderstanding is coming from.

Super AG bonds together two 54mb channels, yes, but it doesn't bond 2.4GHz with 5GHz, it bonds together two of the non overlapping 2.4GHz channels or two non overlapping 5GHz channels.

It cannot operate in two different bands at the same time. There is no such beast. The first AP you linked to has two radios, same as if you put two mini PCI cards in your router.

What you are requesting and complaining a lack of support for does not exist today.

EMP-8601 6G: 802.11a/b/g 316mW
WLM54AG
Sparklan :: WMIA166AG 802.11a/b/g Mini PCI Module

Most people here are familiar with that card. It does 2.4GHz or 5GHz at one time, not both.

hellbound, you are talking about regular turbo which is supported in routeros for a long time. it does not use two modes at the same time, but channels, as 0ldman said. none of the links that you provided do say something about simultaneous use of multiple modes.

i've tried to explain that several times - and ended in accusations of being rude etc. etc. this is USELESS.

see next post - http://forum.mikrotik.com//viewtopic.php?t=12998

i wonder what will be next!

hellbound, you are talking about regular turbo which is supported in routeros for a long time. it does not use two modes at the same time, but channels, as 0ldman said. none of the links that you provided do say something about simultaneous use of multiple modes.

I think you ha missted the simultaneous word!
1. open this > http://www.sparklan.com/product_details.php?prod_id=13
2. press CTRL+F and search for Simultaneous. it will show it to you.

to my understanding :
11g turbo uses 40mhz in 2.4 ghz
11a turbo uses 40mhz in 5.x ghz


11ag, uses 40mhz in 2.4 plus 40mhz in 5.8 doubling actual throughout.
I've seen another orinoco ap-4000 triband which I think also is using the same concept.

in fact, I don't care about this topic anymore, and I don't really care if mikrotik add new features or not, what I need urgently is that mikrotik fixes the bugs in its current so called stable 2.9.38, I have two access point crashing on me with 40 user on each one of them. that I can't even make support file after crash, I'll post my hand phone screen shot in another post.

thanks.

hellbound, you are talking about regular turbo which is supported in routeros for a long time. it does not use two modes at the same time, but channels, as 0ldman said. none of the links that you provided do say something about simultaneous use of multiple modes.

I think you ha missted the simultaneous word!
1. open this > http://www.sparklan.com/product_details.php?prod_id=13
2. press CTRL+F and search for Simultaneous. it will show it to you.

to my understanding :
11g turbo uses 40mhz in 2.4 ghz
11a turbo uses 40mhz in 5.x ghz


11ag, uses 40mhz in 2.4 plus 40mhz in 5.8 doubling actual troughout.
I've seen another orinoco ap-4000 triband which I think also is using the same concept.

in fact, I don't care about this topic anymore, and I don't really care if mikrotik add new features or not, what I need urgently is that mikrotik fixes the bugs in its current so called stable 2.9.38, I have two access poing crashing on me with 40 user on each one of them. that I can't even make support file after crash, I'll post my handphone screen shot in another post.

thanks.

as has been mentioned several times in this thread, the Sparklan AP runs two radios. I have never seen a single radio run multiple frequencies like that. The standard channel width is around 20MHz, turbo uses 40MHz. That is why it has double bandwidth, it has nothing to do with a dual band card. It is available on dual band cards, but does not require them.
Any router or AP that runs 2.4GHz and 5GHz simutaniously is using two radios.

Hellbound,

There are two radios in the sparklan product you linked to above.

You can have same effect in Mikrotik if you put two R52 cards in your RB532. Configure one card for 802.11b/g and the other for 802.11a. Connect a 2.4GHz and 5.8Ghz omnis to the respective card.

Now you have all three bands (a, b & g) in one router. This is basically what sparklan did.

From the link you provided.

SparkLAN WX-7800A Wireless 11a+g Dual-Band Access Point can be accessed in all three 802.11 wireless networks. WX-7800A contains two separate wireless connectivity radio transceivers, which support all three popular wireless networking specifications.

ok guys, you're right about sparklan,

but just one thing I don't understand,
why would Atheros call it Super-AG and
Super-G if Super-AG only operate on
5.X Ghz why not just call it Super-A?

Super AG can work in 2.4GHz or 5GHz, just not both at the same time.

yes, they released this SuperAG when the first a+g cards came out (the ones that can do one OR the other). before them, there were only a or g cards.

what is super?

mikrotik doesn't support all super features?
e.g.
AR=adaptive radio
XR=extended range
Compression (supported)

those are not super features, those are called `marketing tricks`

these days everything is super, ultra, plus, extended, experienced, advanced
if your product does not include one of these in title - you will have unsuccessful product.

other way is to use abbreviations like XP, X, XTR, GT, GTX etc.

all what i ask is - please read documentations, marketing sites very carefully, they want to sell you a product that is no better than others, unless you find real features that is useful for your business.

AFAIK
XR (Extended Range) is a real feature of the chipset.
It allows the radio to receive at lower levels than normal WiFi, probably compromising PER to do it.

XR support has been asked for before - [on:off]

Would be nice if there's a response on that.

Regards

CableFree Solutions

I just hope the buzzword "Xtreme" dies with 2006. I've had enough of "Bearenstien Bears Xtreme Sports" and other silly names.

AFAIK
XR (Extended Range) is a real feature of the chipset.
It allows the radio to receive at lower levels than normal WiFi, probably compromising PER to do it.

XR support has been asked for before - [on:off]

Would be nice if there's a response on that.

Regards

CableFree Solutions

Yes, been requested several times. I would much rather a link slowed to a crawl during extreme fade than dropped altogether. XR can help achieve that.

George

Since knowlege might impart some civility and understanding:

Written only with the intention of helping to reduce the proliferation of misinformation with what I know. The standards evolved out of order, but here goes:

First came Spread Spectrum which could hop across about 80 small channels and they called it FHSS- Frequency Hopping Spread Spectrum with bit rates around possibly around 1Mbps, maybe higher.

Next came 802.11b. This put the signal on the air using a 20MHz wide channel, codified by Complementarty Code Keying to code data symbols into the DSSS-Direct Sequence Spread Spectrum signal. One big wide 'sperad out' signal. Different and higher data rates achieved by 'coding' more data into the 'symbols'. Up to 11Mbps.

Then they figured out it didnt always work, or it needed help with reflections (multipath) caused by obstructions and clutter, and marketing wanted to say it was Non-Line-of-Sight. Engineers finally got mad and told marketing, no its 'Near' Line-of-Sight since it will go through a wall you cant see through, etc.

An innovative engineer figured out that that original FHSS using lots of little channels had its merits. So they invented OFDM. Orthogonal Frequency Division Multiplexing. This concept uses lots of small channels carrying a smaller, but easier to decode amount of data. However, it needed to put all the small channels back together to deliver higher data rates. Simplistically, consider a piano. If you put your arm across the piano keys, the ear can actually hear all the notes. The OFDM reciever can hear all the channels, but it needs to put them back together in the right order because some of them bounce and take a little longer to get there. So the OFDM processor must be fast enough to put them all back together before the 'next' set of data starts coming in. They couldnt put this standard on the 2.4GHz band, because they would never be able to do enough testing to make it work(simple explanation). So they decided to put it on 5GHz and call it 802.11a. This standard had the capability of delivering 54Mbps raw throughput in the same channel size, and had the enhanced characteristic of being able to actually use the bouncing and reflected signals to its advantage and found that it would deliver reliable signals even through cluttered paths, therefore it was NLOS!

Well once they had that 802.11a OFDM worked out, using the same bandwidth, all they had to do then was prove that it didnt trash the band it was operating in, or appear as a significantly louder signal. They measured the average power and other things, and figured the specs to allow the OFDM to operate in the 2.4GHz band. So they had to give it a standard number--802.11g. So the G signal is identical to the A signal.

Once they got all that down, it was somewhat a combination of trickery, engineering and marketing that gives us the rest of the story. Bonding has been around for a long time before wireless, so put two channels together and you might get twice the throughput. However, the wider that signal, the more subject it is to every negative effect that can happen to a digital signal. Now if its Super A, or G, it has to put back together twice the amount of small channels, and the chance that you could lose some of the channels is probably more than 'twice'.

Other engineers figured that why not put more transmitters (in the SAME band) and/or antennas running in sync with each other to gain more throughput. This idea has merit, but the benefit, as it should be, is more reliability, and that is what really gets you the extra throughput.

Further proprietary variations and enhancements to receiver and antenna technology, without affecting the 'standard', also yield better reliability, and ultimately deliver more data. MIMO, XR, beam forming, XSPAN, Adaptive Radio and other magic are all good things.

Finally, the marketing guys figured out that if the engineers developed stuff that worked better, and they did their jobs to get the word out, they wouldnt have to play numbers games to sell more product because the equipment actually would sell itself if it actually did what the specifications said.

One more thing the engineers could do without much problem is make the essentially client-driven, indoor LAN based technology be more service provider driven. 802.11 is a 'CLIENT DRIVEN' protocol and to all the service providers out there, why would you want your 'clients' dictating how you 'the provider' delivers 'ACCESS' to your 'SERVICE?' So some engineers in some companies created software driven methods for the 'service provider access point' to exert control over when the 'clients' radio, or far end of a backhaul circuit, transmits and recieves, and maybe even provide a method to 'tune' parameters to improve reliability. If you have read this far, I appreciate it, and this is why the 'Nstreme' technology doesnt disconnect and reconnect--I am not an expert at this--but it is a result of putting the transmission control back where it belongs. This is not the only way to accomplish this, but it sure helps when attempting to operate a reliable service, and I am just glad to see that someone figured out how to overcome the inherent limitations of the upside down protocol rules.

So, now here comes WiMax. Thats just got to be another thread because its 802.16! BTW, WiMax uses OFDM. Guess who controls the transmission in this protocol, and guess what market segment and application requirement it is designed for?


BTW, I am new to using Mikrotik, and to this forum. What I have seen in this thread is a very enthusiastic group of wireless advocates that are all arguing intensely to accomplish the same end--putting more viable and reliable networks out there that dont break us to operate. I also operate other technologies that are virtually plug and play, but the opportunity to customize, integrate, control, tune to your own liking and for that matter, create 244km links from mountains going over water is a lot more fun than pluggin and playin.

As far as the original question, there is no right answer. My two cents for an answer that would give you 'A/B/G' in a way no-one actually suggested, is use 3 radio cards, three feedlines. 2-Omni Antennas, one 5Ghz physically isolated(as much as possible) directional antenna with as much Front-to-Back and Side-to-Side and Cross-Polarization rejection as possible. Make the 5GHz isolated antenna horizonal polarization, and operate it as low down the band as possible, with the 'A' Access Point as far up the band as possible. Thats all I can come up with for what could be an integrated ABG access point with backhaul......

Thats my story, and I am sticking to it. The verbosity was the only way to completely respond to this wild thread.

Good propagation to everyone...73s

that was cool. thanks for informations

Thanks for reading my long-winded explanation. Makes the pain in my wrist and fingers worth it if it helps. The fact that there are more and more wireless operators out there fighting the telecom gorillas (even though some of them are the actual Internet glue) continues to prove the viability of wireless. This needs to happen constantly. Been doing this a long time, and I still learn a new thing every day. The existence and intensity of this forum is evidence that the number of entrepreneurs out there that want to control their own destiny is growing. It is more difficult, but with a little help from our friends, we can not only get what we need, we can get what we want--because we built it (careful...we asked for it!). Enough of that!

Now I know I need some sleep, the network problems will still be there later this morning....

I am a licensed Electrical Engineer and your description above is one of the best descriptions of the IEEE wireless evolution. It is also a great description of modulation. I am excited to see a more technical approach to answers digging into the physics of radio transmission. So many of the problems WISP operators encounter are RF related and nothing to do with folks at Mikrotik. I want to welcome you to the Mikrotik world. I have been using their RouterOS on both AP, Backhaul and CPE for three years. We have had great success. The Mikrotik community is great and the intellectual benefits are overwhelming. Cisco has a lot to learn. I hope the gentlemen in Latvia continue to evolve and provide great solutions to the rest of the world.

Thanks for the comments, they are appreciated. Building and operating reliable networks requires a diverse skill set.

Fighting the proliferation of misinformation foisted on us by relentless marketing of the 'best' solutions has been a mission of mine for as long as I can remember. No doubt gets me into a lot of 'discussions' and gains me a lot of 'friends'.

Its unfortunate that when a vendor actually does come up with a great solution, certain entities within the space naturally resist. It doesnt surprise me anymore when blanket statements like 'Just doesnt work!' are made. Upon further research, what isnt working usually isnt the equipment. There are a lot of products that work (some not as good as othersl), but the community needs to understand what is 'suitable for a particular purpose'.

I think I have made my point. I just cant wait until the community is forced to deal with WiMax. Explanations of what it can do, and what it is suitable for even evade the WiMax community.

Keep on building reliable networks, and everything will work itself out...73s

Cisco has a lot to learn.

That's not really fair to say that. I believe Cisco has achieved its goal and is the dominant in professional market. they have all the best engineers. simply because they don't want to enter a certain market doesn't mean they don't know how to do it.

Good luck

You are probably right to say that it is not fair. All I am really trying to express is that Cisco needs to foster a more open forum of ideas. They make a great product and they are indeed the standard in IP Routing. Not sure they are the best "Engineers" though.